Document Character Extraction and Normalization for Contract Keywords

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional information processing devices struggle with handling keywords with spelling inconsistencies, making it difficult to manage relevant contract details uniformly.

Innovation Solution

A document processing system that includes a server device capable of extracting and normalizing character strings from documents, such as contracts, to unify keyword representation and display their positions, using techniques like OCR, entity recognition, and Levenshtein distance for spelling correction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If keyword detection is performed on contract text, then important contract parts can be recognized, but spelling inconsistencies cause the detected keywords to be managed differently, making unified management difficult

Engineering Contradiction:

Improvekeyword detection accuracyVSAvoidhandling of spelling variations

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system changes the parameter of keyword representation by introducing normalized keywords that store both the original detected keyword and a standardized form. This allows the system to maintain detection accuracy while adapting to spelling variations through the normalization process, resolving the contradiction between reliable detection and versatile handling.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces an intermediary component (the normalization module with Levenshtein distance calculation) that mediates between raw keyword detection and unified management. This intermediary standardizes keywords with spelling inconsistencies while preserving the original detection results, enabling both accurate detection and unified management.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If all detected keywords are managed separately due to spelling inconsistencies, then each keyword can be tracked individually, but unified management of relevant contract details becomes impossible

Engineering Contradiction:

Improvekeyword tracking precisionVSAvoidunified management capability

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system merges the management of multiple keyword variants by introducing normalized keywords that group together original keywords with spelling inconsistencies. The normalization process combines similar keywords under a unified standard form, enabling ease of operation while maintaining precise tracking through the association between original and normalized keywords.

Inventive Principle:

Principle #5Merging (Combining)

3Productivity

If conventional keyword detection is used without normalization, then the process is simple and fast, but spelling inconsistencies lead to fragmented information management

Engineering Contradiction:

Improvekeyword detection speedVSAvoidinformation fragmentation

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The system performs preliminary normalization action on detected keywords before final management and display. By pre-processing keywords to create normalized forms with spelling correction, the system prevents information fragmentation from occurring, while maintaining productivity through efficient Levenshtein distance calculations and structured storage.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12450938B2Document processing method, and information processing device

Publication Date: 2025.10.21 LEGALON TECHNOLOGIES INC

AI summary

A document processing method comprising: obtaining a character string indicating a content of a document extracting from document information; and obtaining a normalized extracted information by normalizing the character string information in the document information.